Consulting and training for teams that have AI switched on and want it doing more than answering questions.
Most AI in NetSuite today is a chat box answering a question, or a report that used to take a SuiteQL query. That's real, and it's the easy version.
The harder, more valuable version is AI that builds working software and does governed work against live data. It's what I've spent the last two years building for my own practice, and it's what this engagement is about.
Model Context Protocol servers and tools that let an AI agent read and act on NetSuite and application data under role- and scope-based security, so an agent can only see or change what the signed-in user may.
AI agents doing multi-step work — discovery, testing, documentation — with an approval step wherever the result would change something real.
Taking a problem from a written specification to a deployed application, as a web app or a SuiteApp, with an end-to-end test suite generated alongside it.
Requirements interviews and BRDs generated against a leading-practice catalogue, then reviewed and approved by the people who own the process.
Getting reliable output from a model by controlling what it sees, not by writing longer prompts.
Training and change management for consultants and admins, built around how they'll use AI on configuration, testing and documentation in real accounts.
The point of listing these isn't volume. It's that the advice comes from someone who has hit the failure modes in production and designed around them.
The guardrails are the first thing designed, not the last thing added.
Every AI workflow I set up follows the same rule the applications here do. The AI proposes; a named person on your side accepts or rejects; the decision is recorded.
In practice that means scoped access for every agent, nothing written to a live account without a sandbox test first, and an audit trail your controls review can read. It also means pacing: approval only means something if the approver has time to read what they're approving.
Tell me what your team is using it for today, and what you wish it could do. That's where the useful conversation starts.
Get in touch